Files in this item



application/pdfSHIH-DISSERTATION-2017.pdf (25MB)
(no description provided)PDF


Title:Spatial-temporal patterns in evolutionary ecology and fluid turbulence
Author(s):Shih, Hong-Yan
Director of Research:Goldenfeld, Nigel
Doctoral Committee Chair(s):Dahmen, Karin
Doctoral Committee Member(s):Maslov, Sergei; Kuhlman, Thomas
Department / Program:Physics
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):rapid evolution, individual-level model, demographic stochasticity, predator-prey dynamics, master equation, phenotypic fluctuations, collective coevolution, cyanobacteria, cyanophage, Prochlorococcus, range expansion, laminar-turbulent transition, non-equilibrium phase transition, pipe flow, directed percolation, universality class, extreme value statistics, finite size scaling
Abstract:This thesis explores the turbulence of ecosystems, and the ecology of turbulence. Ecosystems and turbulent fluids are both highly non-equilibrium and exhibit spatio-temporal complexity during the course of their evolution. It might seem that they are too complicated to extract universal properties, even if there are any. Surprisingly, it turns out that each of them can shed light on the other, enabling them both to be solved. In particular, the techniques used to explore ecosystem dynamics turn out to be exactly what is needed to solve the problem of the laminar-turbulent transition in pipes. Accordingly, this thesis is organized into two parts. Part 1 discusses what governs the rate of evolution and what are the consequences of the interplay between ecology and evolution at different scales. Three different aspects of these underlying questions are included in this part: (1) We first study the phenomenon of anomalous population dynamics known as "rapid evolution", in which a fast evolutionary time scale emerges from intense ecological interactions between species. Specific examples are rotifer-algae and bacteria-phage, where the ecosystem is composed of a predator and its prey. However, a sub-population of mutant prey arises from strong environmental pressure, and the trade-off between selection from reproduction and predation is manifested in the patterns of eco-evolutionary dynamics. We discuss how to solve such system with inherent stochasticity by a generic and systematic analytical approach in the spirit of statistical mechanics, using a stochastic individual-level model. We show that this method can naturally capture the universal behavior of the stochastic dynamics from demographic noise without any additional and more biologically detailed assumptions. (2) Second, we address the question of the role of selection in evolution and its relationship with phenotypic fluctuations. Phenotypic fluctuations have been conjectured to be beneficial characteristics to protect against fluctuating selection from environmental changes. But it is not well-understood how phenotypic fluctuations shape the evolutionary trajectories of organisms. We address these questions in the context of directed evolution experiments on bacterial chemotactic phenotypes. Our stochastic modeling and experiments on the evolution of chemotactic fronts suggest that the strength of selection can determine whether or not phenotypic fluctuations grow or shrink during successive rounds of selection and growth. (3) The third aspect of the first part focuses on the paradox of coexistent stability in microbial ecosystems that display especially intricate evolutionary phenomena. We propose that horizontal gene transfer, an important evolutionary driving force, is also the driving force that can stabilize microbe-virus ecosystems. The particular biological system for our model is that of the marine cyanobacteria Prochlorococcus spp., one of the most abundant organisms on the planet, and its phage predator. Phylogenetic analysis reveals compelling evidence for horizontal gene transfer of photosynthesis genes between the bacteria and phage. We test our hypothesis by building a spatially-extended stochastic individual-level model and show that the presence of viral-mediated horizontal gene transfer can induce collective coevolution and ecosystem stability, leading to a large pan-genome, an accelerated evolutionary timescale, and the emergence of ecotypes that are adapted to the stratified levels of light transmission as a function of ocean depth. The goal of Part 2 is to understand the nature of the transition to turbulence in fluids, which has been a puzzle for more than a century. The novelty of our approach is that we consider transitional turbulence as a non-equilibrium phase transition. Accordingly we attempt to approach this problem by looking for an appropriate long-wavelength effective theory. We report evidence of candidate long-wavelength collective modes in direct numerical simulations of the Navier-Stokes equations in a pipe geometry, where we uncover unexpected spatio-temporal patterns reminiscent of ecological predator-prey dynamics. This finding allows us to construct a minimal Landau theory for transitional turbulence, which resembles a stochastic predator-prey model. This in turn can be mapped into the generic universality class of directed percolation. Stochastic simulations of this spatial-extended individual-level predator-prey model are able to recapitulate the experimentally observed super-exponential dependence of the lifetime of turbulent regions on Reynolds number near the onset of turbulence. We argue that these remarkable scaling phenomena reflect the presence of finite-size effects as the correlation length becomes of order the pipe diameter, leading to a universal finite-size scaling distribution for the velocity fluctuations.
Issue Date:2017-08-31
Rights Information:Copyright 2017 Hong-Yan Shih
Date Available in IDEALS:2018-03-13
Date Deposited:2017-12

This item appears in the following Collection(s)

Item Statistics